Seeking Visual Discomfort: Curiosity-driven Representations for Reinforcement Learning
This work addresses sample efficiency issues in vision-based RL for control tasks, offering incremental improvements in exploration and representation quality.
The paper tackles the problem of sample efficiency and generalization in vision-based reinforcement learning by improving sample diversity for state representation learning, resulting in boosted visitation of problematic states, improved state representations, and outperformance of baselines across all tested environments.
Vision-based reinforcement learning (RL) is a promising approach to solve control tasks involving images as the main observation. State-of-the-art RL algorithms still struggle in terms of sample efficiency, especially when using image observations. This has led to increased attention on integrating state representation learning (SRL) techniques into the RL pipeline. Work in this field demonstrates a substantial improvement in sample efficiency among other benefits. However, to take full advantage of this paradigm, the quality of samples used for training plays a crucial role. More importantly, the diversity of these samples could affect the sample efficiency of vision-based RL, but also its generalization capability. In this work, we present an approach to improve sample diversity for state representation learning. Our method enhances the exploration capability of RL algorithms, by taking advantage of the SRL setup. Our experiments show that our proposed approach boosts the visitation of problematic states, improves the learned state representation, and outperforms the baselines for all tested environments. These results are most apparent for environments where the baseline methods struggle. Even in simple environments, our method stabilizes the training, reduces the reward variance, and promotes sample efficiency.